1. Field
One embodiment of the invention relates to an information processing apparatus such as a personal computer, and an image processing method used in the apparatus.
2. Description of the Related Art
There have recently been developed personal computers having the same AV functions as those of audio video (AV) devices such as a digital versatile disc (DVD) player and TV apparatus.
A personal computer of this type requires a function of decoding and reproducing image data which is compression-encoded by a block-encoding method such as MPEG. The block-encoding method such as MPEG processes image data block by block. Pixel groups adjacent to a block boundary are continuous on the time base, but belong to different blocks. Hence, these pixel groups are quantized at different precisions. Image data obtained by decoding block-encoded image data readily suffers block noise owing to distortion which occurs near the block boundary.
As a technique of reducing block noise, Jpn. Pat. Appln. KOKAI Publication No. 2002-232889 discloses a technique of calculating a reference correction amount in consideration of the pixel values of four pixels near a block boundary and correcting the pixel values of the four pixels by using the reference correction amount.
In recent years, most deblocking processing circuits for removing block noise adopt a low-pass filter. The low-pass filter smoothes the pixel value of each pixel in a pixel group near a block boundary by using the pixel values of pixels near the target pixel. This can reduce distortion of a signal near the block boundary.
When decoding and reproduction of image data are executed by software in an information processing apparatus such as a personal computer, the calculation amount of deblocking processing must be minimized.
Deblocking processing using a low-pass filter requires a large calculation amount. Hence, it is not practical to apply deblocking processing using the low-pass filter to the personal computer.
If deblocking processing using a low-pass filter is applied to image data containing a high-frequency component near a block boundary, the high-frequency component is lost by smoothing processing by the low-pass filter.